Skip to content
Back to search

think-mcp

MCP Server for reasoning

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/rai220--think-mcp/badge.svg)](/listings/rai220--think-mcp)

HTML

<a href="/listings/rai220--think-mcp"><img src="/api/listings/rai220--think-mcp/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 68 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Think MCP Tool
  2. Think MCP is an implementation of an MCP (Model Context Protocol) server that provides a "think" tool for structured reasoning in agentic AI workflows. This project is inspired by the Anthropic engineering article: [The "think" tool: Enabling Claude to stop and think in complex tool use situations](https://www.anthropic.com/engineering/claude-think-tool).
  3. According to the referenced article, adding the think tool can lead to improved evaluation metrics by enabling reasoning capabilities even in models that do not natively possess advanced reasoning skills.
  4. ![alt text](tau_bench.png)
  5. <a href="https://glama.ai/mcp/servers/@Rai220/think-mcp">
  6. <img width="380" height="200" src="https://glama.ai/mcp/servers/@Rai220/think-mcp/badge" alt="Think Tool MCP server" />
  7. </a>
  8. ## What is the "think" tool?
  9. The "think" tool allows an AI agent to pause and record an explicit thought during complex reasoning or multi-step tool use. It does not change the environment or database, but appends the thought to the log, helping the agent process information, backtrack, or comply with detailed policies.
  10. This approach is especially useful for:
  11. - Tool output analysis (processing results of previous tool calls)
  12. - Policy-heavy environments (verifying compliance with guidelines)
  13. - Sequential decision making (where each step builds on previous ones)
  14. ## Features
  15. - Implements the "think" tool as described in Anthropic's research
  16. - Minimal, standards-based MCP server using [mcp[cli]](https://pypi.org/project/mcp/)
  17. - Ready for integration with Claude or other agentic LLMs
  18. ## Usage
  19. ### MCP server configuration
  20. Add this MCP server to your facorite agent.
  21. ```
  22. "mcpServers": {
  23. "think-mcp": {
  24. "command": "uvx",
  25. "args": ["think-mcp"],
  26. "enabled": true
  27. }
  28. }
  29. ```
  30. ## Tool definition
  31. The "think" tool is defined as:
  32. - **Input:** `thought` (string) — A thought to think about.
  33. - **Behavior:** Appends the thought to the log for structured reasoning.
  34. ## Advanced mode
  35. Adds aditional tools for your agent:
  36. - criticize
  37. - plan
  38. - search
  39. ```
  40. "mcpServers": {
  41. "think-mcp": {
  42. "command": "uvx",
  43. "args": ["think-mcp", "--advanced"],
  44. "enabled": true,
  45. "env": {
  46. "TAVILY_API_KEY": ... YOUR TAVILY API KEY HERE ...
  47. }
  48. }
  49. }
  50. ```
  51. ## Reference
  52. - Based on: [Anthropic Engineering Blog — The "think" tool](https://www.anthropic.com/engineering/claude-think-tool)
  53. ## License
  54. MIT License — see [LICENSE](LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.